The Relationship Between Muscle Protein Content and CT-Derived Muscle Radio-Density In

Version 2: December 2016

The relationship between muscle protein content and CT-derived muscle radio-density in patients with upper GI cancer

1,4Michael I Ramage, 1Neil Johns, 1Christopher D. A. Deans, 1James A. Ross, 2Thomas Preston, 1Richard J. E. Skipworth, 3Carsten Jacobi, 1Kenneth C. H. Fearon†

1 Department of Clinical Surgery, Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh, EH16 4SA

2 Stable Isotope Biochemistry Laboratory, Scottish Universities Environmental Research Centre, Rankine Avenue, Scottish Enterprise Technology Park, East Kilbride, Glasgow, G75 0QF

3 Musculoskeletal Diseases Area, Muscle Group, Novartis Pharma AG, Novartis Campus, WSJ-152.2.72.04, CH-4056, Basel, SWITZERLAND

4 Corresponding Author: (Michael Ramage)

† Deceased

Email addresses: (Michael Ramage), (Neil Johns), (Christopher D. A. Deans), (James A. Ross), (Thomas Preston), (Richard Skipworth), (Carsten Jacobi)
Abstract

Introduction

Cancer cachexia is a multifactorial syndrome characterized by skeletal muscle loss. Cross-sectional analysis of CT scans is a recognized research method for assessing skeletal muscle volume. However, little is known about the relationship between CT-derived estimates of muscle radio-density (SMD) and muscle protein content. We assessed the relationship between CT-derived body composition variables and the protein content of muscle biopsies from cancer patients.

Methods

Rectus abdominis biopsies from cancer patients (n=32) were analyzed for protein content and correlated with phenotypic data gathered using CT body composition software.

Results

Skeletal muscle protein content varied widely between patients (median µg/mg wet weight = 89.3, range 70-141). There was a weak positive correlation between muscle protein content and SMD (r=0.406, p 0.021), and a weak positive correlation between protein content and percentage weight change (r=0.416, p 0.018).

Conclusion

The protein content of skeletal muscle varies widely in cancer patients and cannot be accurately predicted by CT-derived muscle radio-density.

Key Words

Body Composition

Cachexia

Cancer

Imaging analysis

Protein content

Skeletal muscle

Introduction

Cancer cachexia has been defined as a multifactorial syndrome characterized by ongoing loss of skeletal muscle mass that cannot be fully reversed by conventional nutritional support (1). Cachexia affects the majority of patients with advanced cancer and is associated with a reduction in treatment tolerance, response to therapy, quality of life and duration of survival. The assessment and classification of cachexia is important for patient prognostication and treatment guidance. Modern cross-sectional imaging techniques, combined with software planimetry, have been proposed as one method for the assessment of skeletal muscle and fat mass in cachectic patients. In this technique, the anatomical limits of different tissue areas are defined by radio-density (measured in Hounsfield units (HU)). Lean mass (LM) and total fat mass can then be predicted by calculating skeletal muscle and fat areas at the level of the third lumbar vertebra (L3), and extrapolating these values to the whole body by using validated regression equations (2). Increased fat infiltration of skeletal muscle (myosteatosis) results in lower skeletal muscle radio-density (SMD) on CT (3).

For cancer patients, image analysis can be performed on CT scans undertaken as part of the patient’s routine staging investigations. Low skeletal muscle index (SMI, defined as skeletal muscle area normalized for height in m2 (4)), as assessed by CT, has been shown to be associated with poor prognosis in a wide range of solid epithelial malignancies. Moreover, reduced SMD is also thought to be an independent predictor of adverse outcome in respiratory and GI cancer patients (5). However, it is not known whether reduced SMD may also reflect variation in skeletal muscle protein content. We aimed to ascertain whether body composition variables measured by standard L3 CT scan analytic software are related to the measured protein content of skeletal muscle biopsies in humans.

Methods

Patients undergoing surgical resection for upper gastrointestinal cancer (4 gastric, 12 oesophageal, 3 junctional, 12 pancreatic and 1 duodenal) were recruited. Patients were defined as weight-losing if their weight at diagnosis was >5% less than their pre-illness stable weight (1). Sarcopenia was defined using SMI cutoffs according to BMI (4). SMD was measured between -30 and 150 HU, with low SMD defined as <41 HU with BMI <25 and <33 with BMI >25 (4).Routine blood sampling was performed, including plasma C-reactive protein (CRP) levels measured by automated turbidimetry. Preoperative staging CT scans were analyzed at L3 level using semi-automated Slice-O-Matic software v4.2 (Tomovision Montreal-Canada), which defines area measurements of skeletal muscle, subcutaneous fat, and visceral fat.

Under general anaesthesia during resectional surgery, rectus abdominis muscle biopsies were taken, snap-frozen in liquid nitrogen, and stored at -80°C in monitored freezers prior to analysis. These biopsies were pulverized and weighed using an analytical balance (Mettler Toledo), then lysed with Phosphosafe Extraction Reagent (Merck Millipore, Billerica, Mass., USA) before being homogenized and centrifuged. The supernatant was analyzed for total soluble protein (including myofibrillar and cystosolic) content using commercially available BCA protein assay kits (Pierce Biotechnology, Thermo Fisher Scientific, Rockford, IL) (6).

Data were analyzed using descriptive statistics. For variables of interest, correlation analysis was performed by non-parametric Spearman’s rank coefficient, and comparison of groups was by Mann-Whitney U-test using IBM SPSS Statistics (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Microsoft Excel was used for graphical analyses. Ethical approval for this study was granted by the local Research Ethics Committee.

Results

Thirty-two (n=32; M:F 26:6) cancer patients (median age 64.5 years, range 43-83) were included. 17 (53.1%) were weight-stable and 15 (46.9%) were weight-losing. 2 patients had Stage 1 disease, 6 had Stage 2 disease, 15 had Stage 3 disease, 5 had Stage 4 disease, and in 4 patients disease stage was not recorded. By pre-established CT criteria (4), 19 (59.4%) were sarcopenic and 13 (40.6%) were not sarcopenic; 25 (78.1%) had normal SMD and 7 (21.9%) had low SMD (related to myosteatosis).

Table 1

Results are shown in Table 1. Median skeletal muscle protein content (µg/mg wet weight) varied significantly between cancer patients, and was significantly lower in the weight-losing patients compared with the weight-stable group (88.1 range (71-97) versus 92.3 (83-141) p=0.027). There was a weak positive correlation between protein content and percentage weight change (r=0.416, p=0.018) (Fig.1). However, muscle protein content was similar in sarcopenic and non-sarcopenic patients as defined by CT (88.3 (71-141) versus 89.7 (74-128), p=0.79). Muscle protein content was lower in the “low SMD” group than the “normal SMD” group (median 93.1, range 71-141 vs 83.2, 74-90; p=0.03), and there was a weak positive correlation between muscle protein content and SMD (r=0.406, p=0.021) (Fig 1b). Muscle protein content did not correlate significantly with any other CT marker of body composition, including muscle area, stature-adjusted muscle area, visceral adipose tissue area, or subcutaneous adipose tissue area. Additionally, muscle protein content did not correlate significantly with age, weight, BMI, or plasma CRP.

SMD correlated negatively with age (r= -0.543, p 0.001) and visceral adipose tissue area (r=-0.384, p=0.04), but not levels of systemic inflammation as indexed by plasma CRP (r=-0.175, p 0.337), suggesting that myosteatosis increased with age and visceral adiposity.

There was no significant difference in skeletal muscle radiodensity, skeletal muscle index, skeletal muscle protein content, visceral adipose tissue area, subcutaneous adipose tissue area, or percentage weight change across categories of cancer stage.

Figure 1a

Figure 1b

Discussion

In the present study, there was substantial variation in the skeletal muscle protein content of the cancer patients. The median level observed was reduced compared with values used to inform standard reference tables and reference man (7) (average 172µg/mg wet weight; range 166.9 – 177.5; n=3 healthy, young males). As such, the present findings are consistent with previous compartmental body composition of weight-losing GI cancer patients using measurement of whole body nitrogen, potassium and water (8). The latter study identified that, compared with healthy controls, both muscle mass and protein content are reduced by >60% with a coefficient of variation for muscle protein content rising from 31% to 77% in the weight-losing individuals compared with the weight-stable controls (8). Factors that might contribute to a reduced protein content in the wet weight of muscle from cancer patients include increased protein degradation, reduced protein synthesis, increased intramuscular fat (myosteatosis) (3), or the relative expansion of extracellular water space (9). Alternatively, in comparison with reference man, there may be differences based on different methodologies: the colorimetric protein assay used in the present study is sensitive to certain amino-acid residues that may not be present in the same proportion in all tissues and, although precise, may not be as accurate as a physico-chemical assay for nitrogen as an index of protein mass. For example, the BCA technique relies on the reduction of Cu2+ to Cu+ by protein with secondary detection of the reduced copper; this renders it susceptible to the presence of other reductive agents present in the sample. Nitrogen assays, however, will be affected by non-protein nitrogen sources and also by non-muscle protein. Comparative studies in this field have been performed in the food industry but remain to be done in human skeletal muscle.

There was a weak correlation between muscle protein content and the degree of weight loss in the present study (Figure 1b). A reduction in the myofibrillar content of muscle could explain previous studies in cancer patients that have shown a reduction in muscle strength when expressed in relation to the cross-sectional area of the muscle (10). Current intervention studies rely on cross-sectional imaging to look for a positive effect of anabolic agents on muscle, using volume as the index measure, with the assumption that the muscle protein content remains constant. The present results show substantial variation in muscle protein content and suggest that CT-derived muscle volume does not necessarily relate to muscle protein mass. That some weight-stable patients had biopsies that demonstrated a relatively low muscle protein content (Figure 1b) may reflect the concept that in cachexia there may be a pre-cachectic state where metabolic change occurs ahead of weight-loss (1).

Limitations of the present study include the absence of healthy controls, the diverse patient cohort and the relatively small sample size. Also, although the protein extraction method allows measurement of total soluble (including myofibrillar) protein, it may not reflect total cellular protein, as the insoluble pellet remaining is likely to contain polymerized-protein filaments and some nuclear and organelle proteins. Additionally, the analytical CT software produces a mean HU measurement of all skeletal muscle present on a CT scan slice, whereas the biopsies represent only a small portion of one of the many different muscles present at L3.

Whereas weight-losing patients (i.e. cachexia) had much lower protein content than weight-stable individuals a similar pattern was not observed for sarcopenic versus non-sarcopenic individuals. Whilst weight-loss is a dynamic measure, sarcopenia may result from a variety of current and historical events and it would probably require a much larger cohort of patients to dissect out the difference.

Conclusions

The present study suggests that, in cancer patients, muscle protein content varies widely and cannot accurately be predicted by CT image analysis.

Acknowledgements

Thanks go to the Tissue Injury and Repair Group, Novartis, NHS Lothian, and the University of Edinburgh

Statements of authorship

Concept and writing performed by MR, RS, NJ, CD, CJ, TP, JR, and KF

Body composition analysis and interpretation by MR, NJ, TP, RS, and KF

Muscle protein content assayed by CJ

All authors critically revised the article for important intellectual content, and all approved the final submitted version

Conflict of Interest statement and funding sources

This study was supported by Novartis [PO number 3001673636], and by ESPEN [ESPEN Research Fellowship Grant 2015].

CJ is an employee of Novartis.

All authors have completed ICMJE CoI declaration forms.

References

1. Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E, Fainsinger RL, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol. 2011 May;12(5):489–95.

2. Mourtzakis M, Prado CMM, Lieffers JR, Reiman T, McCargar LJ, Baracos VE. A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Appl Physiol Nutr Metab. 2008 Oct;33(5):997–1006.

3. Aubrey J, Esfandiari N, Baracos VE, Buteau FA, Frenette J, Putman CT, et al. Measurement of skeletal muscle radiation attenuation and basis of its biological variation. Acta Physiol Oxf Engl. 2014 Mar;210(3):489–97.

4. Martin L, Birdsell L, MacDonald N, Reiman T, Clandinin MT, McCargar LJ, et al. Cancer Cachexia in the Age of Obesity: Skeletal Muscle Depletion Is a Powerful Prognostic Factor, Independent of Body Mass Index. J Clin Oncol. 2013 Apr 20;31(12):1539–47.

5. Prado CM, Lieffers JR, McCargar LJ, Reiman T, Sawyer MB, Martin L, et al. Prevalence and clinical implications of sarcopenic obesity in patients with solid tumours of the respiratory and gastrointestinal tracts: a population-based study. Lancet Oncol. 2008;9(7):629–35.

6. Johns N, Hatakeyama S, Stephens NA, Degen M, Degen S, Frieauff W, et al. Clinical classification of cancer cachexia: phenotypic correlates in human skeletal muscle. PloS One. 2014;9(1):e83618.

7. Dickerson JWT, Widdowson EM. Chemical changes in skeletal muscle during development. Biochem J. 1960 Feb 1;74(2):247–57.

8. Cohn SH, Gartenhaus W, Sawitsky A, Rai K, Zanzi I, Vaswani A, et al. Compartmental body composition of cancer patients by measurement of total body nitrogen, potassium, and water. Metabolism. 1981 Mar 1;30(3):222–9.

9. Fearon KC, Preston T. Body composition in cancer cachexia. Infusionstherapie Basel. 1990 Apr;63–6.

10. Stephens NA, Gray C, MacDonald AJ, Tan BH, Gallagher IJ, Skipworth RJE, et al. Sexual dimorphism modulates the impact of cancer cachexia on lower limb muscle mass and function. Clin Nutr. 2012 Aug;31(4):499–505.

Figure and Table legends

Figure legend

Figure 1a

Figure 1b

Figure 1a and 1b – Correlation between protein content of rectus abdominis muscle biopsy in 32 patients with upper GI cancer and skeletal muscle radio-density (Fig 1a) or percentage weight loss (Fig 1b). Dashed line represents 5% weight loss.

Table legend

Table 1 – Weight, weight change, CT body composition variables and protein content of rectus abdominis muscle biopsy of 32 patients with upper GI cancer